Re: [guppi-list] Guppi Development Snapshot Released

[Jon Trowbridge <trow emccta com>]

On Fri, Aug 04, 2000 at 03:54:38PM +0200, Cesar Talon wrote:
> When I start guppi, I found it took a lot of time (several minutes) when I
> opened a data file. I've discovered that this is because it scan the full path
> from which the program is launched, if a start it inside a single directory
> (without further ramification) it runs ok.

Isn't it amazing how quick hacks always come back to bite you.

Guppi's hard-wired search paths for glade files has some funny stuff in
it (like "..") that lets it work properly from inside of src/guppi
without having done "make install" --- this is purely for my
convenience during development, so I can test things from right inside
of the source tree.

This is obviously bad if you are sitting in /home/foo when you run,
as you'll end up scanning everything under /home the first time
Guppi looks for a glade file.  I was stupid not to have thought of
that.

Anyway, I'll just have to make my hack more clever... I'll probably do
something like add ".." to the search path only if "guppi.c" is in the
current directory or something like that...

Thanks,
-JT

-- 
GNU/Linux: Free your mind and your OS will follow.